BCaBA cheat sheet

Three-term contingency: motivating operation (MO) alters the value of a consequence and evokes behavior; discriminative stimulus (SD) signals availability of reinforcement; the response produces the consequence (SR).

Reinforcement contingency requires three elements: a response, a consequence delivered contingent on that response, and a resulting increase in future response frequency; delivery without behavior change is not proof of reinforcement.

Positive reinforcement adds a stimulus after behavior; negative reinforcement removes or postpones a stimulus after behavior. Both increase behavior. Punishment (positive or negative) decreases behavior.

Extinction: withhold the maintaining reinforcer. Extinction burst (temporary increase in rate, magnitude, or variability) is expected; spontaneous recovery may occur after a pause.

Schedules: fixed ratio (pause after reinforcement, ratio strain at high values); variable ratio (high steady rate, resistant to extinction); fixed interval (scalloped pattern, post-reinforcement pause); variable interval (moderate steady rate).

Discrimination training reinforces a response in the presence of one SD and withholds it in the presence of others (S-delta), producing stimulus control; stimulus generalization is responding to similar stimuli without direct training.

Shaping reinforces successive approximations to a terminal response; chaining teaches a task analysis link by link (forward, backward, or total task). Backward chaining is useful when the final step is most reinforcing.

Functional behavior assessment methods: indirect (interviews, rating scales), descriptive (ABC recording, scatterplot), and functional analysis (systematic manipulation of antecedents and consequences). Functional analysis provides the strongest evidence of function.

Behavior is defined operationally when it is observable, measurable, and described in a way two observers could agree on; a response class groups topographies sharing the same function.

Measurement: frequency/rate (count per observation time), duration (total time engaged), latency (time from SD or instruction to response onset), interresponse time (time between consecutive responses), and percentage of opportunities.

Continuous recording captures every response; discontinuous methods (partial-interval, whole-interval, momentary time sampling) estimate occurrence and can distort rate-partial-interval tends to overestimate, whole-interval to underestimate.

Interobserver agreement (IOA) formulas: total count IOA = smaller/larger × 100; interval-by-interval IOA = agreements/(agreements + disagreements) × 100; target at least 80% agreement.

Graphing: line graphs plot time on the x-axis and behavior on the y-axis; a level shift, trend change, or variability change across conditions supports a treatment effect within a single-case design.

Reversal (ABAB) design demonstrates experimental control by showing behavior changes when the intervention is introduced, withdrawn, and reintroduced; ethical concerns limit withdrawal of effective treatment.

Multiple baseline design staggers intervention across behaviors, settings, or participants; a change occurring only when intervention is applied to each tier demonstrates control without withdrawal.

Alternating treatments design rapidly compares two or more interventions; it requires discriminable conditions and controls for sequence effects through randomization or counterbalancing.

Changing criterion design applies a gradually shifting performance criterion; each step change in criterion is followed by a corresponding behavior change, documenting a functional relation.

Preference assessment hierarchy: free operant, single stimulus, paired stimulus (forced choice), and multiple stimulus with or without replacement; paired stimulus yields the most consistent rank order.

Reinforcer identification: stimulus preference assessments identify preferred items; a reinforcer assessment then verifies that contingent delivery actually increases behavior.

Establishing operations increase the value of a reinforcer (e.g., deprivation, attention restriction); abolishing operations decrease value (e.g., satiation). MOs also evoke behavior related to the relevant consequence.

Differential reinforcement: DRA reinforces an alternative behavior; DRI reinforces behavior incompatible with the target; DRO reinforces the absence of the target during an interval; DRL reinforces lower rates of the target.

Token economies deliver conditioned reinforcers exchangeable for backup reinforcers; the token must be paired with the backup reinforcer to acquire reinforcing value.

Behavioral contracts specify target behaviors, consequences, and signatures of involved parties; they are most effective when consequences are delivered consistently and immediately.

Prompt hierarchy from most to least intrusive: full physical, partial physical, model, gesture, verbal; most-to-least and least-to-most fading both transfer stimulus control to the natural SD.

Time-out from positive reinforcement requires a reinforcing environment before removal, brief duration, and no attention during time-out; response cost removes a specified amount of a reinforcer contingent on behavior.

Extinction may be combined with differential reinforcement to reduce extinction bursts; the maintaining reinforcer must be identified before extinction is implemented.

Antecedent interventions: noncontingent reinforcement (time-based delivery of the maintaining reinforcer), high-probability request sequence, errorless learning, and priming or precorrection.

Behavioral momentum: presenting several high-probability requests followed by a low-probability request increases compliance with the low-probability request.

Functional communication training (FCT) teaches a socially acceptable response that produces the same reinforcer maintaining problem behavior; it is a DRA procedure.

Behavior reduction plan components: operational definition, baseline data, function-based intervention, reinforcement schedule, crisis plan, and data collection for progress monitoring.

Treatment integrity (fidelity) is the degree to which an intervention is implemented as designed; without adequate integrity, conclusions about effectiveness are invalid.

Social validity addresses whether goals are acceptable, procedures are appropriate, and outcomes are meaningful to consumers, caregivers, and stakeholders.

Generalization strategies: teach multiple exemplars, program common stimuli, use natural contingencies, and reinforce generalized responding (train loosely).

Maintenance is assessed after intervention ends; it is supported by thinning schedules of reinforcement and transferring control to naturally occurring contingencies.

Ethical obligations: obtain informed consent, use least restrictive effective procedures, prioritize client welfare, and ensure services are within the behavior analyst's competence.

Behavior analysts must collect and graph data to justify continuation, modification, or discontinuation of an intervention; subjective impressions alone are insufficient.

Confidentiality applies to records, data, and verbal reports; disclose information only with consent or as legally required.

Supervision and delegation: tasks delegated must be within the supervisee's training and competence, and the supervisor remains responsible for the case.

A risk-benefit analysis weighs potential harm of a procedure against expected benefit; punishment-based procedures require closer monitoring and stronger justification.

Baseline data must be stable or show a predictable trend before intervention begins; an unstable baseline weakens conclusions from a single-case design.

Behavior analysts avoid dual relationships that could impair objectivity and must report conflicts of interest.

Documentation should record objective descriptions of behavior, not labels or inferences, and must be maintained for the required retention period.

Operant conditioning modifies behavior through consequences; respondent (classical) conditioning pairs a neutral stimulus with an unconditioned stimulus until the neutral stimulus elicits a conditioned response.

Reinforcement schedules of the matching law: relative rate of responding matches relative rate of reinforcement delivered across alternatives.

Delay discounting describes the decreased value of a reinforcer as its delay increases; immediate consequences exert stronger control than delayed ones.

Stimulus equivalence: reflexivity (A=A), symmetry (if A=B then B=A), and transitivity (if A=B and B=C then A=C); all three establish derived relations without direct training.

Rule-governed behavior is controlled by verbal descriptions of contingencies; contingency-shaped behavior is acquired through direct exposure to consequences.

Motivating operations are either unconditioned (e.g., food deprivation) or conditioned (e.g., seeing a menu); both alter reinforcer value and evoke behavior.

The Premack principle: a higher-probability behavior can reinforce a lower-probability behavior when access is made contingent.

Data-based decision rules: change the intervention when the trend is opposite the intended direction, when performance is stable and unacceptable, or when variability is so high that no trend is discernible.

Level, trend, and variability are the three dimensions of graphed data used to compare baseline and intervention conditions.

Percentage of nonoverlapping data and visual inspection are common but imperfect measures; statistical methods supplement but do not replace visual analysis.

A functional analysis condition set typically includes alone/ignore, attention, demand, and tangible; elevated responding in one condition identifies the maintaining reinforcer.

Indirect assessments (e.g., functional assessment interviews, motivation assessment scales) generate hypotheses but do not demonstrate function; they should be confirmed by direct observation or experimental analysis.

Preference assessment data are summarized by calculating the percentage of trials or time each item was selected; the highest-ranked item is the presumed reinforcer.

Task analysis breaks a complex skill into observable component steps; data are collected on each step to guide prompting and fading decisions.

Behavioral skills training (BST) sequence: instruction, modeling, rehearsal, and feedback; it is used to teach staff and caregivers to implement procedures with fidelity.

Performance feedback should be specific, delivered close in time to the behavior, and describe both correct and incorrect performance.

Caregiver training requires assessing the caregiver's current skills, setting observable goals, and monitoring implementation in the natural environment.

Collaboration with other professionals requires clear role delineation and sharing of data with consent while maintaining confidentiality.

Cultural responsiveness: assessment and intervention should account for the client's language, values, and family context, and reinforcers should be validated within that context.

Behavior analysts must obtain assent from clients when possible, especially minors or individuals with limited communication, and honor signs of dissent.

If data show the intervention is ineffective, the analyst is obligated to modify the plan rather than continue an ineffective procedure.

Response class hierarchy: members of a response class are functionally equivalent even when topographically different; treatment should target the function, not a single topography.

Setting events (e.g., sleep deprivation, illness) are antecedent conditions that alter the value of consequences and change the likelihood of problem behavior without directly evoking it.

Discriminative stimulus versus MO: the SD signals that a reinforcer is available for a response; the MO changes the value of the reinforcer and the momentary frequency of the behavior.

Conditioned punishers and conditioned reinforcers acquire their function through pairing with unconditioned or established stimuli.

Ratio strain occurs when ratio requirements increase too quickly, producing long post-reinforcement pauses and possibly escape or aggression; gradual thinning prevents it.

Thinning schedules of reinforcement gradually (e.g., increasing ratio or interval requirements) supports maintenance and reduces abrupt behavior change.

A behavior contract or contingency plan should specify the response, the consequence, the schedule, and who delivers the consequence.

Single-case designs require at least three demonstrations of an effect (e.g., three reversals or three tiers) to establish a functional relation.

Baseline logic compares prediction from baseline with verification of the intervention effect and replication across conditions, subjects, or behaviors.

A withdrawal design is contraindicated when the target behavior is dangerous or when withdrawing treatment would be unethical.

In a multiple baseline, tiers should be independent; generalization across tiers can obscure the effect and is a limitation of the design.

Alternating treatments design compares conditions rapidly but requires that the conditions be discriminable and that carryover effects be minimized.

Graphs should include condition labels, phase change lines, axis labels with units, and data points connected within but not across conditions.

Progress monitoring uses frequent, repeated measurement of the target behavior to detect change early and adjust the intervention.

Reinforcer effectiveness depends on deprivation, immediacy, contingency, and magnitude; a reinforcer delivered noncontingently loses its contingency-based effect.

Immediacy: the longer the delay between response and consequence, the weaker the effect; bridging stimuli (e.g., tokens) can mediate delays.

Magnitude: larger or higher-quality reinforcers generally produce stronger responding, but satiation limits repeated delivery of the same reinforcer.

Response effort influences the likelihood of behavior; lower-effort responses are more likely when reinforcement is equal across alternatives.

Choice behavior is influenced by relative rates, delays, and magnitudes of reinforcement rather than absolute values of a single alternative.

Behavioral contrast: when reinforcement conditions change in one component of a multiple schedule, responding in the unchanged component may change in the opposite direction.

The matching law predicts distribution of behavior across concurrently available reinforcement schedules.

Operational definitions should specify the behavior's topography, the context in which it occurs, and examples and non-examples to guide observers.

Measurement reliability improves with observer training, clear definitions, and periodic recalculation of IOA.

Rate is calculated as count divided by observation time; it allows comparison across sessions of unequal length.

Duration recording is appropriate for behaviors that occur for extended periods (e.g., on-task, tantrums); latency recording is appropriate for behaviors that follow a specific stimulus.

Partial-interval recording records whether behavior occurred at any point in the interval; it tends to overestimate duration and is not appropriate for high-frequency brief behaviors.

Momentary time sampling records whether behavior is occurring at the moment the interval ends; it is less biased than partial- or whole-interval recording for estimating duration.

Permanent product recording measures the tangible outcome of behavior (e.g., completed worksheets) and is useful when direct observation is impractical.

Extinction of attention-maintained behavior requires withholding attention, including reprimands, during the extinction phase; other behavior may temporarily increase.

Escape-maintained behavior is reduced by eliminating or reducing demands during extinction and teaching an appropriate escape response (FCT).

Tangible-maintained behavior is reduced by withholding the item during extinction and teaching an appropriate request.

Automatic reinforcement is not socially mediated; extinction may require masking or altering the sensory consequence (e.g., providing competing stimulation).

Punishment procedures (e.g., response cost, time-out, overcorrection) should be used only after less restrictive procedures are insufficient and with appropriate consent and monitoring.

Overcorrection includes restitutional (repair the environment) and positive practice (repeatedly perform the correct behavior) components.

Response cost must be planned so that the individual retains access to reinforcement; removing all reinforcement can produce escape or aggression.

Behavior analysts must use evidence-based interventions and avoid pseudoscientific or unvalidated procedures.

Informed consent includes the purpose, procedures, risks, benefits, alternatives, and the right to withdraw at any time.

Least restrictive alternative: choose the least intrusive effective procedure that addresses the behavior's function.

Behavior analysts must monitor for side effects of intervention, including increases in other problem behavior, and adjust accordingly.

Data should be shared with caregivers in understandable terms, and recommendations should be based on the client's data rather than on generalization from other cases.

When a behavior analyst lacks competence in an area, referral or supervision is required rather than proceeding independently.

Behavior analysts must avoid making claims about outcomes that data do not support.

A functional analysis manipulates antecedents and consequences to isolate the variable maintaining behavior; it is the most rigorous FBA method.

A brief functional analysis uses short sessions and a limited number of conditions to identify function quickly.

A trial-based functional analysis embeds brief test and control trials in the natural environment, improving feasibility in classrooms.

Descriptive assessment (ABC recording) identifies correlations between antecedents, behavior, and consequences but cannot establish causation.

Indirect methods (interviews, questionnaires) are efficient but rely on recall and may be inaccurate; they should be corroborated by direct observation.

The results of an FBA inform the selection of a function-based intervention: attention (reinforce appropriate attention-seeking), escape (teach appropriate escape), tangible (teach requesting), automatic (provide competing stimulation).

Function-based interventions produce larger and more durable reductions in problem behavior than non-function-based interventions.

Chaining methods: forward chaining teaches the first link first; backward chaining teaches the last link first; total task presentation teaches all links each session with prompting as needed.

Task analyses should be individualized and validated by observing a competent performer or consulting the literature.

Prompting should be faded systematically (e.g., most-to-least, least-to-most, time delay) to avoid prompt dependence.

Time delay procedures (progressive or constant) transfer stimulus control from the prompt to the natural SD.

Discrete trial teaching presents a clear instruction, a prompt if needed, a response, and a consequence; it is effective for skill acquisition but should be paired with naturalistic teaching.

Incidental teaching and natural environment training use the learner's motivation to create teaching opportunities within ongoing activities.

Generalization should be programmed from the start by training multiple exemplars and reinforcing in the natural setting.

Confidentiality: records should be stored securely, and identifying information should be removed from data shared for consultation or research unless consent is obtained.

Behavior analysts must comply with legal requirements, including mandated reporting of suspected abuse or neglect.

Multiple relationships that could impair objectivity or exploit the client must be avoided; if unavoidable, disclose and manage the risk.

Supervision should include direct observation, performance feedback, and documentation of supervisee competence.

Behavior analysts must ensure that interventions are implemented with fidelity and that data are collected as specified in the plan.

Ethical decision-making models involve identifying the problem, consulting relevant guidelines and colleagues, evaluating options, acting, and documenting the decision.

Behavior analysts should advocate for clients' access to effective services and avoid abandoning clients without appropriate transition.

Conditioned reinforcement: a neutral stimulus paired with a primary reinforcer acquires reinforcing properties; generalized conditioned reinforcers (e.g., tokens, praise) are paired with multiple primary reinforcers and are less susceptible to satiation.

Schedules of reinforcement can be combined: concurrent schedules (two or more simultaneously available), multiple schedules (alternating with discriminative stimuli), and mixed schedules (alternating without discriminative stimuli).

A multiple schedule uses different SDs to signal different reinforcement conditions; it is used to assess stimulus control and to compare interventions.

A chained schedule requires a sequence of responses, each signaled by a distinct SD, with reinforcement only at the end of the chain.

A concurrent schedule is used to study choice and the matching law.

Differential reinforcement of other behavior (DRO) delivers reinforcement after an interval without the target; it does not teach an alternative skill and should be paired with skill acquisition.

Differential reinforcement of low rates (DRL) is used for behaviors that are acceptable at lower rates, not for behaviors that should be eliminated.

Behavior analysts should use data to evaluate treatment effectiveness and make modifications when the data indicate the plan is not working.

Baseline should be measured long enough to establish a stable pattern; a minimum of three data points is often recommended, but stability matters more than count.

Visual analysis examines level (mean value), trend (slope), and variability (spread) within and across conditions.

A change in level or trend that coincides with the introduction of the intervention, and that replicates, supports a functional relation.

Treatment integrity data should be collected and graphed alongside client data to interpret outcomes accurately.

Social validity can be assessed with questionnaires or interviews with caregivers, clients, and staff.

Maintenance and generalization data should be collected after treatment ends and in untrained settings.

Motivating operations can be unconditioned (e.g., sleep deprivation, hunger) or conditioned (e.g., seeing a preferred item); they alter both reinforcer value and the momentary probability of behavior.

Abolishing operations reduce reinforcer value and decrease the behavior that produces that reinforcer.

Discriminative stimuli do not alter reinforcer value; they signal the availability of reinforcement based on a history of differential consequences.

Stimulus control develops when a behavior is reinforced in the presence of one stimulus and not in the presence of another.

Stimulus generalization occurs when a response trained to one stimulus occurs in the presence of similar stimuli; generalization gradients show the degree of control.

Stimulus equivalence classes can be formed through training a subset of relations; derived relations emerge without direct reinforcement.

Rule-governed behavior can be insensitive to changes in contingencies, which is a limitation when contingencies change.

Preference assessments should be conducted before intervention to identify potential reinforcers; a reinforcer assessment confirms that the item functions as a reinforcer.

Free operant preference assessment allows the learner to engage with items freely and measures duration of engagement.

Paired stimulus assessment presents two items per trial and calculates the percentage of trials each item was selected.

Multiple stimulus without replacement presents an array and removes the selected item; it is efficient but may not detect preferences for items not selected early.

Multiple stimulus with replacement presents an array and replaces the selected item, allowing repeated selection of highly preferred items.

Single stimulus assessment presents one item at a time and records approach or engagement; it is useful for individuals with limited choice-making skills.

Reinforcer assessments should be conducted periodically because preference can change over time.

Behavior reduction should be based on a functional assessment; interventions that do not address function are less likely to be effective.

Antecedent interventions are proactive and include altering the environment, providing choices, and modifying task demands.

Noncontingent reinforcement delivers the maintaining reinforcer on a time-based schedule, independent of behavior, reducing motivation for problem behavior.

High-probability request sequences increase compliance with low-probability requests.

Functional communication training teaches a response that produces the same reinforcer as problem behavior, making problem behavior unnecessary.

Extinction should be implemented consistently; intermittent reinforcement of problem behavior during extinction can strengthen it.

Crisis procedures should be planned in advance and include de-escalation and safety measures.

Behavior analysts must obtain informed consent before assessment and intervention, and must explain procedures in understandable language.

Assent should be sought from individuals who cannot legally consent, and signs of dissent should be honored when possible.

Behavior analysts must protect client dignity by avoiding unnecessary restriction and by using the least restrictive effective procedure.

Behavior analysts must maintain competence through continuing education and supervision.

Behavior analysts must avoid conflicts of interest and disclose any that arise.

Behavior analysts must report data accurately and not fabricate or alter data.

Behavior analysts must follow the law and relevant ethical codes, and consult when uncertain.

Measurement should be selected based on the dimensions of behavior that are relevant to the target (frequency, duration, latency, intensity).

Frequency recording is appropriate for discrete, countable behaviors with a clear onset and offset.

Duration recording is appropriate for behaviors that persist over time.

Latency recording is appropriate when the time between a stimulus and a response is the target.

Interresponse time is the time between consecutive responses and is used to assess response patterning.

Percentage of opportunities is used when the number of opportunities varies across sessions.

Trials-to-criterion is a measure of learning efficiency and is used in skill acquisition programs.

A reversal design requires that the behavior return to baseline levels when the intervention is withdrawn; if it does not, the design is inconclusive.

A multiple baseline across behaviors requires that the behaviors be independent; if they are not, generalization can obscure the effect.

An alternating treatments design requires that the conditions be discriminable and that carryover effects be minimized.

A changing criterion design requires that the criterion changes be gradual and that behavior follow each change.

A multiple probe design uses intermittent probes rather than continuous baseline measurement, reducing assessment burden.

A withdrawal design may be unethical when the behavior is dangerous or when treatment is clearly effective.

Single-case designs require replication to establish a functional relation.

Reinforcement should be delivered immediately, consistently, and contingently to be effective.

Reinforcers should be selected based on preference assessments and confirmed by reinforcer assessments.

Reinforcement schedules should be thinned gradually to promote maintenance.

Differential reinforcement should be used to teach alternative behaviors and reduce problem behavior.

Token economies should include a clear exchange system and a plan for fading tokens to natural reinforcers.

Behavioral contracts should be written clearly and reviewed regularly.

Prompting should be faded systematically to avoid prompt dependence.

Behavior analysts must collect data on the target behavior and on treatment integrity.

Data should be graphed and analyzed regularly to inform decisions.

Interobserver agreement should be calculated for a portion of sessions.

Data should be shared with caregivers and other professionals as appropriate.

Behavior analysts must maintain records securely and for the required period.

Behavior analysts must evaluate social validity of goals, procedures, and outcomes.

Behavior analysts must monitor for side effects and unintended consequences.

Functional analysis conditions: alone (automatic), attention (social positive), demand (social negative), and tangible (access to item).

A control condition (play) provides noncontingent attention and no demands, serving as a comparison.

Elevated responding in one condition relative to control identifies the maintaining reinforcer.

Functional analysis should be conducted by trained personnel with safeguards for dangerous behavior.

Indirect and descriptive assessments generate hypotheses that should be confirmed by functional analysis when feasible.

Function-based interventions should match the identified function.

Functional analysis results should be used to design reinforcement-based interventions first.

Behavior analysts must ensure that interventions are within their competence and scope of practice.

Behavior analysts must obtain supervision when working outside their competence.

Behavior analysts must avoid making false or misleading statements about their qualifications or services.

Behavior analysts must ensure that advertising is accurate and not misleading.

Behavior analysts must protect the confidentiality of clients in public communications.

Behavior analysts must obtain consent before recording or sharing client information.

Behavior analysts must follow ethical guidelines when conducting research with human participants.

Behavioral skills training (BST) is used to teach staff and caregivers to implement procedures.

BST components: instruction, modeling, rehearsal, and feedback.

Feedback should be specific, immediate, and delivered positively.

Treatment integrity should be monitored and graphed.

Caregiver training should include goals, procedures, and data collection.

Behavior analysts should collaborate with other professionals and share data with consent.

Behavior analysts should evaluate the cultural appropriateness of interventions.

Stimulus equivalence requires reflexivity, symmetry, and transitivity.

Reflexivity: A=A. Symmetry: if A=B then B=A. Transitivity: if A=B and B=C then A=C.

Equivalence classes can be taught by training a subset of relations.

Derived relations are not directly reinforced but emerge from training.

Stimulus equivalence is relevant to teaching language and academic skills.

Generalization should be programmed to promote transfer across settings and people.

Maintenance should be assessed after intervention ends.